In a recent webinar organized by OCLC, Laura Dawson presented a very simple, yet essential definition of metadata: “Metadata is data about data.” In this equation, the latter ‘data’ refers specifically to the content of a pBook or eBook. The former ‘data’ is a collection of certain pieces of information about the content that we use to categorize, place, shelve, describe, identify, locate, etc.
Thankfully, there was no limit imposed on how much of this former ‘data’ there could be. And, while many publishers agree on set pieces of metadata, albeit in a de-facto manner, this need not be the case. Expanding on Laura’s definition, let’s look at some examples of items that could be used effectively within the metadata mix, creating a system of layered metadata from within.
In-line code: Code within the XML structure of content is going to be a very important place to lace strategy into bookselling. The more online discoverability maintains and expands its role in sales of book products, the more publishers are going to have to get smart about the search game. Some publishers have begun to do this already, others are exploring their options, but ultimately, it will become essential for publishers to include advanced SEO techniques in both their content itself as well as the coded wrapper it resides in. From finding hiding places to secret keywords, to tagging content properly and effectively, to refining titles and subtitles outwardly, to changing how things are done on copyright pages and TOCs, in-line code for both pBooks and eBooks is perhaps the new marketing plan. If you are not doing this, start now.
Descritive copy: While our conception of metadata has expanded to be included in what we consider ‘marketing,’ the reverse movement has happened less so. Descriptive copy, being actual marketing copy, which feeds out to online accounts, via ONIX or other delivery method, is indeed part of a book’s metadata proper. It describes the content in a longer form than tags and codes, provides an inticing snippet to the consumer, and represents the whole of the product. This copy, like in-line metadata, needs to be optimized for search, as it will appear many places across the internet driving up the book’s collective page rank. Descriptive copy, while having a duty of its own, also needs to play a role behind the scenes in improving discoverability through search. The more keywords a publisher can integrate and the more search-friendly a publisher can make this copy the greater the benefitw will be for our products in the long run.
Website SEO: Any search optimization performed on web properties related to each book should be related and integrated with search optimization perfomed on the content itself. With the ability to load keywords into XML codes, and, eventually, into the text itself, a strong SEO strategy needs to move beyond the content that we have, beyond the descriptive copy, and onto the streets, so to speak. At the very least, the web team that is setting up websites, landing pages, author blogs, and eCommerce pages must converse with teams that are performing keyword development for editorial and marketing purposes.
Social media: Social media outlets are wonderful for interacting with customers and peers alike, but despite the dissemination of huge amounts of information online, few have begun talking about SEO as it realtes to social media. It is important to note that because social media outlets allow for us to talk about our books, they are additional areas for dessimination of metadata. On the various platforms used, user profiles are highly trafficked, bios are read by many, tweets, syndicated blog posts, messages, and answers to questions are spit out by the million every day. It’s time to get smart about these missives and start loading them with keywords, pushing our SEO strategy to the next level. If every book had, say, 50 keywords that must be included in as many places as possible, social media marketing teams would be hardpressed not to use them thus drawing more attention to a book’s presence. It will not be long before we start using ‘heavy tweets’ in our daily marketing activities.
Paid Search/SEM: If you are engaging in SEM as well as SEO, it remains important to integrate any paid campaigns with the non-paid ones. Good paid search marketers can tell you exactly what people are typing into Google and how many of them are clicking on a link to your books. This information can and should be used by the rest of the team to develop keywords. Here, it is essential to remember that the conversation is two-way. Making sure all of these layers sync up is more important, I believe, than simply performing the tasks of implementing each. To simply make insular teams responsible for each of these items is to miss the point. There is more metadata than we think, beyond the ISBN, and the tags, and the copy, and the pub date, and the format, and the page count, and the price. There is a whole world of metadata out there that we can exploit. The nature of the internet today is that there is no ‘one-stop shop’. There is only multiplicity.
There is one last piece of metadata that I wanted to mention, something that someone perhaps much smarter than I will figure out:
User-gen: Perhaps the most difficult area to navigate, because we have little to no control, user generated content, like customer reviews, is also part of a product’s metadata. It is data about the data. Simply because we do not create it, nor what much control over it’s distribution, does not make it any less effective in describing, identifying, and categorizing our content. In fact, even the smallest items can be of important. Let’s say a certain user is highly active at leaving review on a certain site. Other users who may become familiar with these people may, eventually, begin looking for or even SEARCHING for specific user names in order to follow reviews from a trusted source. With sites like Shelfari, GoodReads, and other sharing sites and options being built into eBooks, user generated content and metadata is going to become perhaps more voluminous that publisher generated metadata. How, in the future, are we going to be able to leverage this data?